Welcome to arraycontext’s documentation!

GPU arrays? Deferred-evaluation arrays? Just plain numpy arrays? You’d like your code to work with all of them? No problem! Comes with pre-made array context implementations for:

• numpy

• pyopencl

• jax.numpy

• pytato (for lazy/deferred evaluation)

• Debugging

• Profiling

arraycontext started life as an array abstraction for use with the meshmode unstrucuted discretization package.

Contents

• The Array Context Abstraction
  • Freezing and thawing
    • Usage guidelines
    • What does this mean concretely?
  • The interface of an array context
  • Types and Type Variables for Arrays and Containers
  • Internal typing helpers (do not import)
  • Canonical locations for type annotations
• Implementations of the Array Context Abstraction
  • Array context based on pyopencl.array
  • Lazy/Deferred evaluation array context based on pytato
    • Compiling a Python callable (Internal)
  • Array context based on jax.numpy
  • numpy coverage
    • Array creation routines
    • Array manipulation routines
    • Linear algebra
    • Logic Functions
    • Mathematical functions
• Array Containers
  • Serialization/deserialization
  • Context retrieval
  • MultiVector support
  • Canonical locations for type annotations
  • Containers with arithmetic
  • Containers based on dataclasses
  • Traversing containers
    • Traversing decorators
    • Freezing and thawing
    • Flattening and unflattening
    • Numpy conversion
    • Algebraic operations
• Other functionality
  • Metadata (“tags”) for Arrays and Array Axes
  • ArrayContext-generating fixture for pytest
  • Program creation for loopy
• Installation
• User-visible Changes
  • Version 2021.1
• Licensing
• Acknowledgments
• 🚀 Github
• 💾 Download Releases

Indices and tables

• Index

• Module Index

• Search Page